Dave Brown wrote: > Hey All, > > Having an interesting problem with a FC6 server regarding host bus > errors - every now and again in the messages log there are the following: > > Apr 20 04:58:05 lftvm01 kernel: ata2.00: exception Emask 0x0 SAct 0x0 > SErr 0x0 action 0x0 > Apr 20 04:58:05 lftvm01 kernel: ata2.00: (BMDMA stat 0x6) > Apr 20 04:58:05 lftvm01 kernel: ata2.00: cmd > 25/00:f8:d7:ef:61/00:00:1c:00:00/e0 tag 0 cdb 0x0 data 126976 in > Apr 20 04:58:05 lftvm01 kernel: res > 51/84:a7:28:f0:61/84:00:1c:00:00/e0 Emask 0x20 (host bus error) > Apr 20 04:58:05 lftvm01 kernel: ata2.00: configured for UDMA/133 > Apr 20 04:58:05 lftvm01 kernel: ata2.01: configured for UDMA/133 > Apr 20 04:58:05 lftvm01 kernel: ata2: EH complete > Apr 20 04:58:05 lftvm01 kernel: SCSI device sdc: 625142448 512-byte hdwr > sectors (320073 MB) > Apr 20 04:58:05 lftvm01 kernel: sdc: Write Protect is off > Apr 20 04:58:05 lftvm01 kernel: SCSI device sdc: write cache: enabled, > read cache: enabled, doesn't support DPO or FUA > Apr 20 04:58:05 lftvm01 kernel: SCSI device sdd: 625142448 512-byte hdwr > sectors (320073 MB) > Apr 20 04:58:05 lftvm01 kernel: sdd: Write Protect is off > Apr 20 04:58:05 lftvm01 kernel: SCSI device sdd: write cache: enabled, > read cache: enabled, doesn't support DPO or FUA > > and then every now and again > Apr 20 04:54:32 lftvm01 kernel: ata2.00: exception Emask 0x0 SAct 0x0 > SErr 0x0 action 0x2 frozen > Apr 20 04:54:32 lftvm01 kernel: ata2.00: cmd > 25/00:08:ff:3d:8c/00:01:1d:00:00/e0 tag 0 cdb 0x0 data 135168 in > Apr 20 04:54:32 lftvm01 kernel: res > 40/00:c7:d0:61:84/84:00:1d:00:00/e0 Emask 0x4 (timeout) > Apr 20 04:54:39 lftvm01 kernel: ata2: port is slow to respond, please be > patient (Status 0xd0) > Apr 20 04:55:02 lftvm01 kernel: ata2: port failed to respond (30 secs, > Status 0xd0) > Apr 20 04:55:02 lftvm01 kernel: ata2: soft resetting port > Apr 20 04:55:02 lftvm01 kernel: ata2.00: configured for UDMA/133 > Apr 20 04:55:02 lftvm01 kernel: ata2.01: configured for UDMA/133 > Apr 20 04:55:02 lftvm01 kernel: ata2: EH complete > Doesn't always cause a crash but looks like sometimes it might. Have had > a look around at various places and haven't found a definitive cause let > alone an answer. > > If anyone has any ideas or theories please throw em out there. Curiously > these errors only occur on ata2, we have 2 drives sitting on ata1 > without an issue. ata1 drives are the "os" and ata2 drives are the "data" > > Some more detail about the machine / OS. > Kernel: 2.6.20-1.2933.fc6PAE > Drives: 4x WD 320Gb SATA Drives > ATA Module: ata_piix > CPU: Dual-Core Intel Xeon 3.0Ghz > Mem: 6Gb > Not exactly sure what the motherboard is - it's a "whitebox" server not > Tier1 (IBM,HP,Dell) You said, "Have had a look around at various places..." but you didn't say what you looked at. Is it always ata2.00 that presents the problem? Have you swapped cables around? It is a good idea to tell folks what you done so when they respond you don't have to say..."tried that and it wasn't it". :-)
